Efficiency Discussions
Contents:
- Electric Lamps (a brief history since Edison)
by martin-h-at-mail.utexas.edu (Martin Harriman) (Tue, 27 Jun 1995)
- Electric Lamps
by huntley-at-ix.netcom.com (WRIGHT HUNTLEY ) (Tue, 27 Jun 1995)
- Aquatic Plants Digest V1 #151
by Keith Rogers <krogers-at-canopus.sim.es.com> (Wed, 28 Jun 1995)
- Electric Lamps
by ac554-at-freenet.carleton.ca (David Whittaker) (Wed, 28 Jun 1995)
- Thomas Alva Edison's trivia page
by martin-h-at-mail.utexas.edu (Martin Harriman) (Thu, 29 Jun 1995)
- Thomas Alva Edison's trivia page
by ac554-at-freenet.carleton.ca (David Whittaker) (Thu, 29 Jun 1995)
- Lumens per watt-a non-scientific observation
by ljw-at-voyagerco.com (Larry West) (Mon, 26 Feb 1996)
- Fluorescent vs. Metal Halides
by PacNeil-at-aol.com (Tue, 27 Feb 1996)
- Lumens per watt
by Dave Gomberg <gomberg/wcf.com> (Wed, 25 Nov 1998)
- CF lamps
by Zxcvbob/aol.com (Fri, 12 Nov 1999)
- Is this enough light?
by busko/stsci.edu (Ivo Busko) (Mon, 31 Jan 2000)
by martin-h-at-mail.utexas.edu (Martin Harriman)
Date: Tue, 27 Jun 1995
Here's a very brief rundown on lamp types for the curious:
Electric lamps can be divided into incandescent (conventional and
quartz-halogen) and discharge. Incandescent lamps (of either variety) put
most of their energy out in the infrared, and so are of little interest
here (think of them as tiny expensive electric space heaters that happen to
light up).
Discharge lamps include fluorescent, mercury, metal-halide, low-pressure
sodium, and high-pressure sodium. They all produce light by passing a
current through a gas, exciting the odd electron here and there. The
electrons emit light as the return to their ground state, and this directly
or indirectly is the source of the illumination.
Fluorescent lamps use low-pressure mercury, which emits primarily up in the
ultraviolet, and (except for blacklight and germicidal lamps) fluorescents
are coated internally with phosphors that glow under ultraviolet. The
visible light they produce is almost all from these phosphors. Some types
of fluorescents make excellent light sources for planted tanks.
Mercury lamps increase the pressure, which shifts more of the direct output
into blue and green emission lines (these can be supplemented with
phosphors to produce better color rendering); these are the blue-green
lights you see illuminating parking lots and the like. These are useless
as grow-lamps; neither you nor your plants' chloroplasts would be thrilled.
Metal-halide lamps introduce metal halides into the mercury lamp; this adds
all sorts of nice emission lines to the spectrum, and greatly improves
color rendition. These, too, make fine light sources for plants (and
people looking at plants).
Low-pressure sodium lamps are not seen much, at least in the United States.
They are the most efficient light source, because they put almost all
their energy into one yellow-orange emission line (and the human eye is
pretty sensitive there). Unfortunately, you can't see colors under this
illumination; everything looks like shades of grey illuminated by a fairly
ghastly yellow light. I have no idea off-hand what the average chloroplast
would think, but you certainly wouldn't like looking at plants under this
light, so the point is moot.
High-pressure sodium lights are very widely used in the United States for
street-lighting. If it is bright, kind of yellow with purply overtones,
and you can pretty much distinguish color under it, it is high-pressure
sodium. GE, at least, has some "white" sodium lamps that actually render
colors fairly well (but nowhere near as well as MH or fluorescent), so
perhaps someday these will make reasonable light sources for planted tanks
(but I don't think "someday" is today).
Right now, compact (biaxial) fluorescents on electronic ballasts can be
both more efficient and better at rendering colors than metal-halide lamps
(in the wattages used to light tanks, at least; I haven't bothered to try
to calculate what you'd need in fluorescent lamps to equal MH lamps over
300 watts). Metal-halide lamps are more compact than fluorescents, though.
As several people have pointed out, you can buy all of this much more
inexpensively from a lighting distributor than from a pet store.
Incidentally, if you decide to go with MH lamps, I would suggest choosing
lamps designed for the particular burning position of your fixtures; these
have very significant efficiency and lifespan advantages over "universal"
lamps. You can get horizontal, vertical base-up, and even vertical
base-down lamps. Be sure, too, that your fixtures are enclosed, so that
arc-tube or envelope ruptures will neither shower your tank with hot
fragments, nor expose you and your tank to short-wave UV (some of the MH
lamp failure modes are a bit dramatic, but a good fixture will be designed
to contain the disaster).
If there's interest, I can supply a more detailed summary of the different
lamp types (fluorescents, in particular, come in a zillion or two different
configurations). Perhaps, though, this has already been done (I didn't see
it in the Krib, but then I'm notorious for missing the obvious).
Incidentally, if you are reading this on a CRT, you might like to consider
the impact ambient light has on that activity, too. Discharge lamps on a
conventional ballast are horrible illumination (they flicker at almost the
same rate as your CRT, most likely), but also extremely common (especially
if you are reading this at work, you naughty person you). On the other
hand, compact biaxial fluorescents on an electronic ballast are pretty
good; they don't flicker much at all, and what fluctuation they have
probably won't interact noticeably with your CRT's flicker. You can even
find them (complete with fancy phosphor and electronic ballast) at
supermarkets sometimes (GE sells some designed to replace incandescent
lamps; they are slightly larger than some of the incandescents they
replace, though).
--Martin Harriman
martin-h-at-mail.utexas.edu
by huntley-at-ix.netcom.com (WRIGHT HUNTLEY )
Date: Tue, 27 Jun 1995
>Electric lamps can be divided into incandescent (conventional and
>quartz-halogen) and discharge. Incandescent lamps (of either variety) put
>most of their energy out in the infrared, and so are of little interest
>here (think of them as tiny expensive electric space heaters that happen to
>light up).
Snip...
Whoa! The rest of this posting was superb, and I thank you for it.
Unfortunately, I must quibble with the above statement. Like incandescent
lamps, the sun is a hot incandescent source, and the best imitator we have
of sunlight is plain old halogen lamps! [I admit the heat often must be
vented]
Plants LIKE deep red light, with their main absorption peak at 670nm and a
weaker one in the violet at about 450 nm. An equivalent blackbody at
2800-3000K is just about right. Most of the gas discharge lamps put out a
line here and a line there, with little regard for the plants' needs. "Cool"
and "Warm" flourescents have gobs of green, right in the chlorophyl
absorption dip, and almost none where plants need it (red and violet). It
makes them "look" bright (i.e., have high Lumens) but they are terrible
plant bulbs, compared to "Daylight" or Chroma 50s. A broad-band incandescent
source is of great interest here, and I have several tanks whose plants can
prove it.
Last December, I reported on the compact-base screw-in flourescents made by
LOA and GE that used tri-phosphors. They have been terrific for plants, too,
but my local stores are now only carrying an inferior brand that uses the
nearly worthless green-rich phosphors. Reliability has been superb, with
only one infant mortality (happily replaced by Home Depot). For highly
illuminated small aquaria, they are very hard to beat.
I hope my one tiny quibble does not detract from an otherwise wonderful
posting. The reef folks have gotten us so off into duplicating skylight that
we sometimes forget the sources our FW plants need.
Wright
by Keith Rogers <krogers-at-canopus.sim.es.com>
Date: Wed, 28 Jun 1995
It was a good mailing, but I'm going to quibble with the quibble. It
is simplistic to look at the spectral absorption of chlorophyll a
disolved in ether (what you see in every intro biology and aquarium
plant book) and make that the sole basis of determining what spectrum
to select. Things have been evolving to capture light from Sol for at
least 3.5 billion years and there's more to it than the intro biology
book would indicate. There are potentially dozens of photoreactive
molecules in a given autotroph, and none of them operate in isolation
in a container of ether. To get an idea about how huge this topic is,
just go to any decent sized university's stacks on photochemistry -
you'll find literally meter after meter of specialty books on the
subject.
Since everything has evolved for our sun the ideal thing to do is to
mimic its output, i.e. a nearly perfect blackbody radiator of ~5600
kelvins -at- whatever energy rate. That's impractical to do artificially
so we're left with mixing and matching bulbs to get enough of what's
important for plants. There are many ways to achieve this and any
reasonable mix of tri-phosphor/daylight/full-spectrum bulbs works well
enough to keep you pruning your tanks every week. What you end up
with depends upon what's available locally, what plants you have, your
personal retnal approval, etc. My experience is that anything that
looks sun-ish and sufficiently bright to people works for the plants
if you're meeting their other needs and that it's easy to get too
up-tight about this lighting thing...
Happy pruning to all,
Keith Rogers
krogers-at-es.com
by ac554-at-freenet.carleton.ca (David Whittaker)
Date: Wed, 28 Jun 1995
>Low-pressure sodium lamps are not seen much, at least in the United States.
>They are the most efficient light source, because they put almost all
>their energy into one yellow-orange emission line (and the human eye is
>pretty sensitive there).
I was under the impression that HPS lamps were the most efficient.
>High-pressure sodium lights are very widely used in the United States for
>street-lighting. If it is bright, kind of yellow with purply overtones,
>and you can pretty much distinguish color under it, it is high-pressure
>sodium. GE, at least, has some "white" sodium lamps that actually render
>colors fairly well (but nowhere near as well as MH or fluorescent), so
>perhaps someday these will make reasonable light sources for planted tanks
>(but I don't think "someday" is today).
The colour temperature of the colour enhanced HPS lamps is in the
2500K range with a CRI of 85. The Son Agro, an Iwasaki horticulture
lamp has a CRI of 21 and a colour temperature of 2000K. If I were
to just grow plants and not care too much about their appearance,
I'd employ regular HPS on the basis of cost and energy efficiency.
The Son Agro is twice as efficient as the colour improved lamp,
and twice as expensive as a regular HPS lamp.
>Incidentally, if you decide to go with MH lamps, I would suggest choosing
>lamps designed for the particular burning position of your fixtures; these
>have very significant efficiency and lifespan advantages over "universal"
>lamps. You can get horizontal, vertical base-up, and even vertical
>base-down lamps.
Universal lamps burn 25% longer and 25% more efficiently in the
vertical position. The problem that one encounters with respect
to lamps engineered to burn horizontally is that they are made to
fit in a special notched socket, which Hamilton etc. do not provide
with their products. If one is building one's own hood one could
look to such a design. Just what I've read.
Dave
- --
by martin-h-at-mail.utexas.edu (Martin Harriman)
Date: Thu, 29 Jun 1995
I got a few questions whose answers I thought might be of general interest.
So here goes:
i) Efficiency champs. Keep in mind that all lumens are not precisely
created equal, of course, but:
conventional incandescent: 500PAR64, 6500 (initial) lumens, 500 watts,
13 lumens/watt
halogen: 50PAR30 (GE Halogen IR), 1000 lumens, 50 watts,
20 lumens/watt
fluorescent T12, GE Chroma 50, 1870 lumens, 40 watts,
46.75 lumens/watt (on a ballast that only draws 40 watts, heigh ho)
fluorescent T12, GE rare earth, 2960 lumens, 40 watts,
74 lumens/watt (but this depends on the ballast)
fluorescent T8, GE rare earth, electronic ballast (!), 2650 lumens, 27 watts,
98 lumens/watt
fluorescent compact F40/30BX/IS, electronic ballast, 2840 lumens, 30 watts,
95 lumens/watt
MH, E28 universal burning horizontally, 7100 lumens, 175 watts,
41 lumens/watt
MH, E28 universal burning VBU, 8300 lumens, 175 watts,
47 lumens/watt
MH, E28 horizontal [EP39 base], 11300 lumens, 175 watts,
65 lumens/watt
MH, E37 VBU, 31000 lumens, 400 watts,
78 lumens/watt
HPS, GE Deluxe Lucalox, 9135 lumens, 150 watts,
61 lumens/watt
HPS, GE White Lucalox, 4160 lumens, 95 watts,
44 lumens/watt
LPS, 19140 lumens, 135 watts,
142 lumens/watt
ii) Magic exploding instant suntan fire-bombs. MH bulbs can fail
catastrophically. Thus, you need to use them in an enclosure capable of
retaining 1100 degree C fragments. Moreover, the enclosure should shield
the unwary from shortwave UV, as these lamps radiate very strongly in this
band if their outer envelope fails. I quote the warning from 21 CFR
1040.30: "This lamp can cause serious skin burn and eye inflammation from
shortwave ultraviolet radiation if outer envelope of the lamp is broken or
punctured, and the arc tube continues to operate. Do not use where people
will remain for more than a few minutes unless adequate shielding or other
saftey precautions are used." MH lamps are safe (no, really!) as long as
they are properly enclosed. Other things to be aware of: the lamps should
be turned off at least once per week for at least fifteen minutes. You
should relamp your fixtures before the end of rated life, since failure and
catastrophic failure probability go up markedly, and light production
declines precipitously past rated life.
iii) Horizontal vs. universal lamps.
Yes, horizontal lamps have a special base. VBU oriented lamps do not (but
the smallest VBU lamp is 400 watts, and it sounds like most people are
using 175 watt pendants).
I hope this is helpful; several people sent me e-mail on these points, so I
thought there might be some interest in the list as a whole.
--Martin Harriman
martin-h-at-mail.utexas.edu
by ac554-at-freenet.carleton.ca (David Whittaker)
Date: Thu, 29 Jun 1995
>i) Efficiency champs. Keep in mind that all lumens are not precisely
>created equal, of course, but:
>
Try 110 lumens/watt for the Iwasaki HPS lamps.The GE DeLuxe and
GE White may be colour improved lamps which are less efficient
I believe.
> LPS, 19140 lumens, 135 watts,
> 142 lumens/watt
>iii) Horizontal vs. universal lamps.
>Yes, horizontal lamps have a special base. VBU oriented lamps do not (but
>the smallest VBU lamp is 400 watts, and it sounds like most people are
>using 175 watt pendants).
Venture makes both horizontal and VBU lamps in 175 watt sizes.
VBU is 67 lumens/watt.
Dave
by ljw-at-voyagerco.com (Larry West)
Date: Mon, 26 Feb 1996
Hi!
I've been reading the debate over lumens, watts, MH vs FL, and thought a
little research was in order. In about 5 minutes, I came up with the
following table from data in Grainger's 385 (1994) catalog:
Ranked in order of efficiency (more lumens per watt):
6: 60 Watt Daylight (6500K) HO Flourescent T-12 (all FLs are 48")
Initial Lumens: 3600
Lumens per watt: 60
5: 110W Cool White (4100K) VHO FL T-12
Initial Lumens: 6950
Lumens per watt: 63.18
4: 40W DAylight (6500K) FL T-12
Initial Lumens: 2600
Lumens per watt: 65
3: 160W Daylight (6500K) VHO FL T-12 72" (the smallest Daylight VHO in the
catalog)
Initial Lumens: 11300
Lumens per watt: 70.625
2: 175W Metal-Halide
Initial Lumens: 13000
Lumens per watt: 74.29
1: 32W TL-80 4100K FL T-8
Initial Lumens: 3050
Lumens per Watt: 89.0625
Outside of our purpose:
180W Low Pressure Sodium (LPS)
Inital Lumens: 33000
Lumens per watt: 183.33
So, it looks like the T-8s have it! Of course, getting enough of them over
a tank can be problematic. It also shows that watts per gallon or any other
reference that uses watts can be misleading, and is probably bordering on
useless (like the much misused and utterly useless 'candlepower'). Lumens
should be a more accurate measurement, and obviously the depth of the tank
is more significant than the capacity.
What surprised me was the inefficiency of the HO & VHO lamps. Of course,
they may be the only solution in getting an acceptable quantity of lumens
into the tank in some situations.
It would be interesting to determine how many lumens we are putting into
our tanks.
Larry West
by PacNeil-at-aol.com
Date: Tue, 27 Feb 1996
In a message dated Mon, 26 Feb 1996 , Larry West writes:
>4: 40W DAylight (6500K) FL T-12
>Initial Lumens: 2600
>Lumens per watt: 65
>3: 160W Daylight (6500K) VHO FL T-12 72" (the smallest Daylight VHO in the
>catalog)
>Initial Lumens: 11300
>Lumens per watt: 70.625
Compact fluorescents fall in between these two:
13W Daylight (5000K)
Initial Lumens: 900
Lumens per watt: 69.231
Just thought I'd through in my $.02 worth.
PacNeil-at-aol.com[Neil Schneider]
by Dave Gomberg <gomberg/wcf.com>
Date: Wed, 25 Nov 1998
OK, so it's not PAR. If you have PAR figures, please add them. Use a
fixed pitch font for best results in viewing.
Incandescent
Vlt Watt Lumens/watt (initial)
120 100 17
Quartz-halogen
12 5 10
10 11
20 15 (35 w/reflector)
35 40
120 20 14
35 17
50 16
100 20
Fluorescent T-12
120 80
Fluorescent T-8
120 90
Fluorescent T-5
120 75 (70 for multiple u-tubes)
Metal halide
120 any 85
Around here, power costs about $.40 per watt-year (based on 3000 hours on
time).
DIY fluorescent costs about $1/watt for nice looking, $.16/watt for a shop
light hanging from the ceiling on chains. Low-voltage halogens cost about
$.53/watt according to Tom Barr.
So per 1000 lumens (and remember, we should be talking about per PAR):
Buy Run($/year)
Incandescent 32 24 (either regular or quartz-halogen)
Fluorescent 11 5 (includes metal halides too, close enough)
I find it very hard to understand how anyone can argue for incandescent as
less costly.
Did I screw up one or more of the computations?
- --
Dave Gomberg, San Francisco mailto:gomberg@wcf.com
http://www.wcf.com/wcf
- -----------------------------------------------------------------
End of Aquatic Plants Digest V3 #667
************************************
To unsubscribe to aquatic-plants, send the command:
unsubscribe aquatic-plants
in the body of a message to "Majordomo@ActWin.com". Archives are
available on the web at http://www.actwin.com/fish/aquatic-plants
or via FTP to ftp.actwin.com in /pub/aquaria/aquatic-plants.
by Zxcvbob/aol.com
Date: Fri, 12 Nov 1999
Steven Dixon <stdixon@ben.bechtel.com> sez:
> Just to make a stupid observation of my own, all seem to agree that the new
> T-8 bulbs are more efficient, which I take to mean that T-8 bulbs put out
> more light per watt (again, I'm not sure if we mean lumens or PAR when this
> is said) than the old T-12 bulbs. So perhaps simply saying that all of
> these bulbs use the same technology doesn't really answer the efficiency
> question.
>
> Does anyone have a definitive answer to this question? Do CF bulbs put out
> more light per watt (lumen or PAR) than regular fluorescent bulbs; than T-8
> bulbs? Do we have data on this point?
I have manufacturer's data (Philips's Lamp Specification and Application
Guide), but I dunno how much info you really want. All of these specs will
be with the new 835 phosphor because it is available in all sizes. The 840
and 850 phosphors produce the same lumens per watt as 835. Linear T5 lamps
have an efficiency of up to 105 lumens/watt, but I've never actually seen one.
F40T12 lamps in the SPEC35 series produce 3200 lumens (80/w) and the 34w
"econowatt" equivalent produces 2800 lumens (82/w). This is a lot better
than the old "cool white", which I can't find good data for anymore for 4'
lamps.
The F15T8 lamps in SPEC35 produces 940 lumens (63/w), so you can see the
affect of the shorter arc distance. The same F15T8 in the old "cool white"
produces 870 lumens (58/w).
The new T8 lamps are better. The F32T8 with 835 coating gives 2950 lumens
(92/w), and there are premium lamps available that do 3200 lumens, for that
magic 100 lumens per watt.
Compact fluorescents fall in-between the T12's and new T8's. The little 13W
with a GX23 base gives 825 lumens (63/w). The long tube CFL's do a lot
better. The 36W lamps that are 16 inches long produce 2900 lumens (80.5/w).
The 40W lamps that are 22" long actually use 38w and produce 3300 lumens
(87/w). The 50W lamps produce 4300 lumens (86/w). My book does not list the
55W nor the 96W CFL's.
Hopefully you can extract something useful from all this.
regards,
bob
by busko/stsci.edu (Ivo Busko)
Date: Mon, 31 Jan 2000
"Roger S. Miller" <rgrmill@rt66.com> wrote:
<snip>
>Ivo Busko might have something more to say about this, but I think if you
>were making your comparison based on PAR rather than lumens you might find
>that the variations are not as great.
<snip>
I agree 100%. For the NO bulbs listed in the table at
http://www.aquabotanic.com/lightcompare.htm, lumens/watt span a range of
about 300%, while PAR/watt only 25%. In other words, different types of
fluorescent phosphors may look very different to us in terms of perceived
brightness, while plants don't "see" too much of a difference between then.
- -Ivo Busko
Baltimore, MD
The Krib
Lighting
The Krib